Oof, are you sure? Facing $9bn damages, Google asks Supreme Court to hear Java spat Google has taken the years-long spat with Oracle over its use of Java code in the Android mobile operating system to the US Supreme Court. The search engine giant announced yesterday that it had asked the US's highest court to review the copyright dispute, which has gone on for the best part of nine years. Google is facing as …
At least the lawyers are making out well. I am sure a few of them are even good people. :)
I find your faith in human^W lawyer nature... touching. Any shreds of goodness, compassion and empathy are slowly tortured out of them at Law School.
After all, it's no coincidence that a lot of lawyers become politicians.
You must be a cynic. :-)
First both companies have deep pockets. At least deep enough to push this all the way to the supreme court.
So if this were a David vs. Goliath I would agree with you... but its not.
Google got caught with their hand in the cookie jar.
To sum it up... they didn't want to pay Oracle for licensing Java ME. Of course when Sun came up with their licensing plan, they didn't envision a phone or other mobile device would be capable of running the full JVM.
If you followed the court case it was a willful infringement. The question really boils down to if they did enough to skirt the law. I'm not going to play lawyer and to be fair... both sides have merit in their position. If SCOTUS decides not to get in to it... Oracle wins. If they do... its going to be a toss up and nobody can predict how it will end.
The smart money is on Oracle because the odds are in their favor.
"The smart money is on Oracle because the odds are in their favor."
I don't think it's an odds game at the moment. Both sides will come in with their stories, the stories will be disputed and attacked from the other side and one story will end up appearing stronger. Until those stories are told and we see how well they are told, I don't think we can predict which way it will go.
Looking purely at the facts, I suspect the odds favour Oracle (i.e. Google probably didn't produce a clean room Dalvik as required) but I'm not sure Oracle have the solid proof required to win this case and some of Oracles previous arguments may not stand up to more serious judicial appraisal which the Supreme Court will provide.
Just to restate again - the case will likely be decided on legal technicalities of what is and isn't presented and accepted by the courts rather than the historical events that lead to Dalvik.
Assume all outcomes weigh the same.
Oracle wins because Google has only a 25% chance when you consider .5 chance of hearing the case and .5 chance of winning. That's 1 in 4.
That said...
I agree the Odds are in Oracle's favor; However, IIRC the less than clean Dalvik build was already part of the case early on and have no real bearing on the facts that the SCOTUS is being asked to hear.
The underlying argument Google is trying to make is that Oracle doesn't have the right to copyright APIs and that it will cause harm if this is allowed to stand. That said, they are also raising the issue of an old case which some courts follow and others do not. (IIRC it was about a spreadsheet compatibility mode) SCOTUS heard that case but split 4 to 4 so no decision was rendered. IMHO its a bit more of a stretch.
This is a bit different and even under the copyright laws, you can still have interoperability under fair use. That said, Google is damned by their own internal emails which show intent to circumvent Sun's initial ME licensing by creating their own VM jump starting it by using Sun's JVM and using the APIs to not pay Oracle ME's licensing fees which are set under FRAND rules.
This is Google's last Hail Mary attempt of not getting forced to payout Billions. (8+ Billion)
BTW, w.r.t odds, you never know how the court will respond. I've personally seen Judges do some pretty dumb things.
Hmm, you speak like you know something about the case but your comments suggest otherwise. JavaME was never an option as it was way too limited for an OS like Android. It was designed as a mobile OS for the Nokia style phones of the time not for access to underlying hardware. It was made for a very small footprint, low resource device and limited use.
The licencing required was for the Java Compatability testing which meant you could call it Java but wasn't available for mobile devices (that had to be JavaME).
So a clean room version based upon harmony was created, the on clean room bit was a tiny amount that was written in by someone who had originally written it and chevked ranges. The rest was the API and created the specification for the interface which was mainly similar as nearly all APIs are to allow interoperability. For instance you have standardised Java Docs, getters and setters etc. But the underlying processing was clean.
Sun seemed perfectly happy with it, Oracle saw a money grab bought Java and sued.
Oracle could be sued for their use of SQL for exactly the same arguments.
Copyright for artistic works never sits great with software development as every single software developer has copied for their projects. And every one who has reused an API of any size has probably done similar to this.
The bit about suing for SQL was slightly tongue in cheek however it definitely has similarities. The SQL invented by IBM may not look much like the SQL of today but it looked similar enough to Oracles first RDBMS in it's constructs, language, calls and structure.
If you equate SQL language to an API then Oracle reused most of the original API and extended it. So what if IBM decided not to develop it and the fact that copying and extending other's work especially when done as a research project was okay.
Hence why SQL became a fairly open standard (ANSI standard that is).
I know as much about this case as the next man.
And suggest you try to find the licensing agreements for Java (SE and ME)
Yes JavaME is a subset that was meant to run on mobile devices at the time. Of course there's Moore's law where you could run Java SE on a mobile device at the time of the lawsuit and even more so today.
But again you're missing the point.
Where Java ME comes in to play was the fact that Google wanted to circumvent payment of licensing fees to Oracle. (Considering the Billions made on Android...)
There's more and the bottom line is that before you start to discuss it... SCOTUS has to first agree to hear the case. Assuming RBG is no capable of hearing the case... you again could find a 4:4 split. Of course that assumes that SCOTUS will consider this along the same lines of the earlier spreadsheet argument which split the court 4:4
The point is that the bottom line is very much in Oracle's favor if they don't screw up in the event the courts do want to hear it.
NOTE: I think both companies are piles of evil shit. I'm merely commenting on it from a legal perspective.
Under Copyright Law, you have fair use which would allow you to use the APIs for interoperability. But in this case... that wasn't the point. Even in the earlier case where SCOTUS split 4:4 the use of interoperability was to steal customers away. So you could see today's SCOTUS even take a more considervative approach and rule in Oracle's favor. (Which is why they will probably agree to not hear the case because it will set a precedence which could have negative ramifications and by not hearing it... the lower courts would have latitude to make their own decisions.)
"Yes JavaME is a subset that was meant to run on mobile devices at the time. Of course there's Moore's law where you could run Java SE on a mobile device at the time of the lawsuit and even more so today.
But again you're missing the point."
WTF? What has Moore's Law got to do with what you can license for a mobile device?
The fact is that JavaME wasn't not close to being suitable for a Smartphone. It was not possibly to actually licence 'Java' for an OS like Android. Clean room implementations were the only way that it was possible or create a new language. As Sun was not legally bothered about Harmony and later endorsed Android then it seemed like a sensible option.
Lest not forget that Oracle originally sued over patents which were all dismissed and then brought copyright into it which became their main lawsuit. After the only copied code found was for 7 lines of a range check and the API, they only had the API left to sue on.
"You must be a cynic. :-)First both companies have deep pockets. At least deep enough to push this all the way to the supreme court.
So if this were a David vs. Goliath I would agree with you... but its not."
:-)
Yeah, I wasn't making a comment about this case specifically, just the general trend for the constant appeal back and fore, which -- as you say -- is unfair when the wealth of the 2 sides isn't equal.
You can't appeal unless you can pick up on a potential error in the applicability of the law that could reasonably change the outcome.
i.e. if the opposition had had a wateright case in the first place, there wouldn't be room for an appeal or the appeal would be dismissed.
It means Google has something which it believes is in error - most likely in the amount of damages awarded, which is one of the reasons that exaggerating damages claims (even if you were in the right) is a bad idea generally... the other side need only prove you were taking the mick to end up costing you even more money.
With something like this going up to higher courts, it's literally saying "the court was in error doing what it did". Not "we don't have a case, your Honour". The Supreme Court might laugh that out, or send it back to the lower court again saying "No, this part isn't right, do it again".
Endless appeals only come about from not securing your case in the first place (which is hard with a complex case, admittedly) and/or by playing off favourable courts against your opponent and not considering that they could argue that later (in which case it would be Oracle that were taking the mick, not Google).
However you paint it, Oracle suffering $9bn of damage because Google made a Java-alike solely for it's Android OS is ridiculous. Especially when they have since gone through not only Dalvik but another two formats of executables that aren't Java-related at all. i.e. there's more than one way to skin a cat, and we can do them all without affecting anyone, and we don't "need" Java technology at all, not even for Java-like Dalvik.
Chances are it'll go through the Supreme Court, who'll limit the damages, and Google won't even pay half of that.
Uhm No.
You are right that Google has to have a reason for the appeal.
Their argument is that there was an earlier case over spreadsheets where SCOTUS split and the lower court's ruling stood, however not all courts agreed with it so they think that this case would be a good precedence and that if they believed in the lower court's ruling ... SCOTUS should overturn the lower court in this case.
That said... Google's argument is weak.
IMHO with the conservative bent in this SCOTUS... if they decide to hear it.. it will still go against Google.
"Unless the Supreme Court corrects these twin reversals, this case will end developers' traditional ability to freely use existing software interfaces to build new generations of computer programs for consumers."
===
will be one of the few times you hear ANY of the big boys menton the 'users' / consumrs
next up, I am expecting - "won't anyone think of the children" :oP
While I think Google's argument is probably correct, I would feel slimy rooting for them. And rooting for Oracle? EWWWW.
While I agree that both companies should lose for well... being Oracle and Google... one has to win.
Google's argument is wrong and a stretch.
Even with copyrights you have fair use. What Google did was not fair use. By their own admission it was to avoid paying Oracle royalties for Sun's JVM.
This isn't actually a meritless case on either side.
Google asks... can you copyright an API.
Oracle says that they stole their IP by using the API among other things in creating their copy of the JVM because they didn't want to license Sun's Java ME. ( And there are internal Google emails detailing that there were conversations over this... )
There is a lot more to this case...
IMHO both companies lack ethics and morals. They are both evil. That said, this case? Oracle has the edge in the lawsuit. The irony? You can partially blame Eric Schmidt since he was at Sun during the creation of Java and its business model and licensing strategy for Java.
This is not good.
Interoperability is at the heart of the technology industry. If APIs and interfaces can be copyrighted, then things such as Wine and other programs can be sued out of existence. Another one that comes to mind is MS-DOS vs. DR-DOS. And then what about UEFI? Let's not forget OpenOffice/LibreOffice's ability to read and write Microsoft Office documents.
The bottom line is that implementation of APIs and interfaces are copyrightable, but the APIs and interfaces themselves should not be.
"The bottom line is that implementation of APIs and interfaces are copyrightable, but the APIs and interfaces themselves should not be."
Exactly. However, any jurisdiction that disagrees will not be too troubled by the argument either way. It's domestic software industry will disappear in double-quick time and the judges' only contact with "software" will be in the form of shrink-wrap packages written abroad.
Clearly there is more to this case than what is being written in the article.
But lets clear some misconceptions on your part.
"Interoperability is at the heart of the technology industry. "
Where is there interoperability between Java and Google's virtual machine?
The point is that Google wrote their version of a JVM as a way to skirt Sun's Licensing Terms.
(This is actually a fact that was brought up during the case.)
The key for Google was to create a virtual machine similar to Java so that it would be adopted by developers. One of Oracle's arguments is that this isn't interoperability but Intellectual Property theft.
The other example you raised.... OpenOffice/LibreOffice's ability to read and write Microsoft Office documents.
This isn't an API issue. The document is written to a published specification. As is UEFI. (Unified Extensible Firmware Interface (UEFI) is a specification for a software program that connects a computer's firmware to its operating system (OS). ) [Google it]
Interoperability by following a published Specification is not illegal nor a violation of copyright laws.
And that's key. Google did not start their design from scratch.
To be clear, I don't want to sound like I am taking sides. However... Oracle's arguments do have merit and there is a lot of misconceptions on the part of everyone. The odds are in Oracle's favor. For google to win, they have to first get SCOTUS to hear the case. They may decide not to hear it. Then if they do decide to hear it, Google has to win. Straight odds? Oracle 75%, Google 25%
With 8.8 Billion and growing on the line? I'd spend a million or two on lawyers to get it in front of SCOTUS too.
While your argument makes sense, that leads to another question: Where does the line get drawn?
Virtually everything is derived from previous stuff. Java itself has copied tons of library code from older languages like C.
Maybe Google created an alternate JVM out of whole cloth, mirroring the existing Java APIs. Big whoop. Java is not original either. It copied tons of syntax, libraries, etc, from languages before it, like C++.
I mean, what next? Is Oracle going to sue Microsoft because C# bears a lot of similarities to Java?
There is way too much mushy middle in this case that seriously threatens the entire computer industry.
This isn't an API issue. The document is written to a published specification. As is UEFI. (Unified Extensible Firmware Interface (UEFI) is a specification for a software program that connects a computer's firmware to its operating system (OS). ) [Google it]
I see what you did there...fnar, fnar!
Shoulda coulda woulda...
You are applying development nicieties to a corporate legal pissing match.
Lawyers are cunts, greedy cunts. They dont care if the c64 is better than the spectrum.
They do care if someone can pay their fees.
There is a lesson here for companies - always check the license before committing the company to a plarform n toolset.
Anyone using .net is going to find this out sooner rather than later.
"If APIs and interfaces can be copyrighted"
They can be, and this has always been the case, as the court ruled. But the doctrine of Fair Use allows use of something even if it is copyrighted. If I interpret the situation correctly (IANAL), Fair Use allows use of APIs and interfaces. And, IIRC, Google was found guilty of copying some implementation, which would not be part of the API.
"the free and open Java language (*)"
(*) terms and conditions apply
This case is also about misrepresentation, or even entrapment. Both Sun and Oracle heavily promoted Java as "open". But in the way Oracle have approached this case, it could not be considered "open" in any conceivable sense of the word.
Google built an entire Java-based ecosystem from the ground up, using only the APIs as the definition of how it should interact. If Oracle have copyright on this part, then the entire language is closed and proprietary, almost by definition. Not only that: any Java program that you write, which *consumes* that API, is subject to Oracle control.
Than to hear it and find for Oracle. If they refuse cert then there's no precedent set across all courts, and even in that circuit they can view it pretty narrowly. If the Supreme Court hears it and rules in favor of Oracle, not only will it apply to all lower courts across the country, but depending on how they word their decision it could end up being read pretty broadly - perhaps broadly enough that theoretically if there are still a few undead cells left in SCO's dried up husk they could take another crack at Linux.
I do NOT want to risk even the slightest chance they find in favor of Oracle, so I really really really hope they refuse to hear Google's appeal.
"The other issue.. suppose they hear it and Google wins?
Now you have no protection for IP in the software industry."
No, if Google wins you have a clear and firm position on what is an interface and what can be copied or not. Like, say, an auto spares company creating parts that are plug in compatible with the original ones. Or a printer ink manufacturer creating cartridges that can be used in printers from other brands. Or a battery plug connector that can be used on many brands of batteries. Or (gasp) a power plug connector that can be used in many countries, or.... the list is endless.
The debate of whether interfaces can be subject to IP property is as young as software itself, because in the past nobody ever even thought of copyrighting the shape of shift gears, or power plugs, or the diameter of tubes, or the shape of threads on a bolt, or... anything at all.
If Oracle wins the entire software industry will flock to open specifications for everything over time, just to avoid the risk of accidentally infringing some interface IP. This will kill IP property in software, forever.
That's a quality FUD out there.
To put it into real world example you seem like so much equivalent of this lawsuit would be as if someone created exact copy of patented PCB but claimed it's a totally different product because they used capacitors from a different manufacturer.
Better if they refuse certThan to hear it and find for Oracle.
If they refuse cert, then they are implicitly agreeing with the decision in favour of Oracle as it exists now. Since the decisions being appealed are from the Federal Circuit, refusing cert means that the FC's decision is the binding precedent for all courts that would use the FC as their appeals court - unless the FC decides to overturn its own precedent (which is possible, but unlikely). And my understanding (IANAL) is that the FC is the appeals court for all Federal Courts on copyright and patent cases.
Therefore cert plus a finding from the Supremes against Oracle is needed to prevent the Oracle-backing rulings from becoming the de-facto precedent across all patent and copyright cases across the US.
Is this apparent hatchetwombling, designed neither by agency nor by deity, to redress the recent advantages awarded the developer versus middle management, or are we all fervently waiting for the first Kubernetes deployment touting a hammer-and-tongs CMS treewise based upon NLP and thus dedicated simply to Liver Herpes Froth?
This is what Fuchsia is all about! If they lose the case, it's unlikely they will be able to continue publishing Android versions, so a successor to Android becomes a completely rational thing to do.
I'd been wondering why they were working to replace the leader of the market, considering it's still at the top of its game... Silly me.
Please please please please deny Google review. I just want to see one of these mega cases finally reach the point when no more appeals are possible, no more legal moves or tricks can happen, and Google literally have to wire transfer 9 billion dollars to Oracle by midnight, or start having their executives arrested.
Yep. According to the original verdict which allowed copyrighting the APIs, that's exactly right. Note that I do not defend said original verdict, but if it stands then so does the second verdict.
Also, note that I specifically mentioned that *using* APIs was perfectly valid (after all, that's what they're there for). So yes, Oracle can call AWS to do work until such a time as they create their own cloud. What they are not allowed to do (according to the first verdict) is create APIs in their product which mimic the AWS's API calls.
Oracle have their own cloud, and they front it with Amazon's S3 API so that customers can move to them without changing anything. Oracle are even suing people for *not* choosing their cloud solution!
Oracle are money-grabbing hypocrites. You clearly have no idea how common a practice re-implementing an API is in IT, nor why vendor lock-in is bad for everyone.
You are welcome to your opinion, but when it is based on ignorance it's value is nil.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
